Insurance-based Disparities in Outcomes and Extracorporeal Membrane Oxygenation Utilization for Hospitalized COVID-19 Patients.

BACKGROUND: The objective of this study was to examine insurance-based disparities in mortality, nonhome discharges, and extracorporeal membrane oxygenation utilization in patients hospitalized with COVID-19. METHODS: Using a national database of U.S. academic medical centers and their affiliated hospitals, the risk-adjusted association between mortality, nonhome discharge, and extracorporeal membrane oxygenation utilization and (1) the type of insurance coverage (private insurance, Medicare, dual enrollment in Medicare and Medicaid, and no insurance) and (2) the weekly hospital COVID-19 burden (0 to 5.0%; 5.1 to 10%, 10.1 to 20%, 20.1 to 30%, and 30.1% and greater) was evaluated. Modeling was expanded to include an interaction between payer status and the weekly hospital COVID-19 burden to examine whether the lack of private insurance was associated with increases in disparities as the COVID-19 burden increased. RESULTS: Among 760,846 patients hospitalized with COVID-19, 214,992 had private insurance, 318,624 had Medicare, 96,192 were dually enrolled in Medicare and Medicaid, 107,548 had Medicaid, and 23,560 had no insurance. Overall, 76,250 died, 211,702 had nonhome discharges, 75,703 were mechanically ventilated, and 2,642 underwent extracorporeal membrane oxygenation. The adjusted odds of death were higher in patients with Medicare (adjusted odds ratio, 1.28 [95% CI, 1.21 to 1.35]; P < 0.0005), dually enrolled (adjusted odds ratio, 1.39 [95% CI, 1.30 to 1.50]; P < 0.0005), Medicaid (adjusted odds ratio, 1.28 [95% CI, 1.20 to 1.36]; P < 0.0005), and no insurance (adjusted odds ratio, 1.43 [95% CI, 1.26 to 1.62]; P < 0.0005) compared to patients with private insurance. Patients with Medicare (adjusted odds ratio, 0.47; [95% CI, 0.39 to 0.58]; P < 0.0005), dually enrolled (adjusted odds ratio, 0.32 [95% CI, 0.24 to 0.43]; P < 0.0005), Medicaid (adjusted odds ratio, 0.70 [95% CI, 0.62 to 0.79]; P < 0.0005), and no insurance (adjusted odds ratio, 0.40 [95% CI, 0.29 to 0.56]; P < 0.001) were less likely to be placed on extracorporeal membrane oxygenation than patients with private insurance. Mortality, nonhome discharges, and extracorporeal membrane oxygenation utilization did not change significantly more in patients with private insurance compared to patients without private insurance as the COVID-19 burden increased. CONCLUSIONS: Among patients with COVID-19, insurance-based disparities in mortality, nonhome discharges, and extracorporeal membrane oxygenation utilization were substantial, but these disparities did not increase as the hospital COVID-19 burden increased.

Low Preoperative Serum Creatinine is Common and Associated With Poor Outcomes After Nonemergent Inpatient Surgery.

OBJECTIVE: To assess the association between low preoperative serum creatinine and postoperative outcomes. BACKGROUND: The association between low creatinine and poor surgical outcomes is not well understood. METHODS: We identified patients with creatinine in the 7 days preceding nonemergent inpatient surgery in the American College of Surgeons National Surgical Quality Improvement Program database from 2005 to 2020. Multivariable logistic regression was used to examine the association between creatinine and 30-day mortality and major complications. RESULTS: Of 1,809,576 patients, 27.8% of males and 23.5% of females had low preoperative serum creatinine, 14.6% experienced complications, and 1.2% died. For males, compared with the reference creatinine of 0.85 to 1.04, those with serum creatinine ≤0.44 had 55% increased odds of mortality [ adjusted odds ratio (aOR), 1.55; 95% CI, 1.29-1.86] and 82% increased odds of major complications (aOR, 1.82; 95% CI, 1.69-1.97). Similarly, for females, compared with the reference range of 0.65 to 0.84, those with serum creatinine ≤0.44 had 49% increased odds of mortality (aOR, 1.49; 95% CI, 1.32-1.67) and 76% increased odds of major complications (aOR, 1.76; 95% CI, 1.70-1.83). These associations persisted for the total cohort, among those with mildly low albumin, and for those with creatinine values measured 8 to 30 days preoperatively. CONCLUSIONS: A low preoperative creatinine is common and associated with poor outcomes after nonemergent inpatient surgery. A low creatinine may help identify high-risk patients who may benefit from further evaluation and optimization.

Perioperative Neurological Evaluation and Management to Lower the Risk of Acute Stroke in Patients Undergoing Noncardiac, Nonneurological Surgery: A Scientific Statement From the American Heart Association/American Stroke Association.

Perioperative stroke is a potentially devastating complication in patients undergoing noncardiac, nonneurological surgery. This scientific statement summarizes established risk factors for perioperative stroke, preoperative and intraoperative strategies to mitigate the risk of stroke, suggestions for postoperative assessments, and treatment approaches for minimizing permanent neurological dysfunction in patients who experience a perioperative stroke. The first section focuses on preoperative optimization, including the role of preoperative carotid revascularization in patients with high-grade carotid stenosis and delaying surgery in patients with recent strokes. The second section reviews intraoperative strategies to reduce the risk of stroke, focusing on blood pressure control, perioperative goal-directed therapy, blood transfusion, and anesthetic technique. Finally, this statement presents strategies for the evaluation and treatment of patients with suspected postoperative strokes and, in particular, highlights the value of rapid recognition of strokes and the early use of intravenous thrombolysis and mechanical embolectomy in appropriate patients.

The Unequal Burden of COVID-19 Deaths in Counties With High Proportions of Black and Hispanic Residents.

BACKGROUND: Understanding the current burden of coronavirus disease 2019 (COVID-19) deaths in vulnerable populations will help inform efforts by policymakers to address disparities in COVID-19 outcomes. OBJECTIVE: The objective of this study was to examine the association between COVID-19 deaths and the county-level proportions of non-Hispanic Black and Hispanic residents. RESEARCH DESIGN AND METHODS: A retrospective study using COVID-19 mortality data from USA Facts linked to data from the US Census Bureau, the Health Resources & Services Administration Area Health Resources file, and the US Census Bureau. Negative binomial regression was used to estimate the association between the total county COVID-19 deaths during consecutive 30-day intervals and the proportion of non-Hispanic Blacks and Hispanic residents after adjusting for resident demographics, comorbidity burden, rurality, social determinants of health, and health care resources. RESULTS: In April, counties (n=179) with >40% Blacks had 6-fold higher death rates than counties (n=1521) with <2% Blacks [incident rate ratio (IRR)=6.58, 95% confidence interval (CI): 3.29-13.2, P<0.001]. These counties had higher death rates until October, but were no different than referent counties in November. In April, death rates in counties with >40% Hispanic residents were similar to death rates in counties with <2% Hispanic residents. Death rates in these counties peaked in August (IRR=3.14, 95% CI: 1.69-5.82, P<0.001) but were also no different than referent counties in November. These effects were robust after adjusting for county-level characteristics. Before August, death rates differed little by insurance status, but since then, counties with >15% uninsurance rates had up to 2-fold higher mortality rates (IRR=1.97, 95% CI: 1.19-3.27, P<0.001) than counties with <5% uninsurance rates. CONCLUSION: Counties with high concentrations of non-Hispanic Blacks were disproportionately affected by COVID-19 throughout most of the pandemic, but other social determinants of health such as health insurance are now playing a more prominent role than race and ethnicity.

Levels of Evidence Supporting the North American and European Perioperative Care Guidelines for Anesthesiologists between 2010 and 2020: A Systematic Review.

BACKGROUND: Although there are thousands of published recommendations in anesthesiology clinical practice guidelines, the extent to which these are supported by high levels of evidence is not known. This study hypothesized that most recommendations in clinical practice guidelines are supported by a low level of evidence. METHODS: A registered (Prospero CRD42020202932) systematic review was conducted of anesthesia evidence-based recommendations from the major North American and European anesthesiology societies between January 2010 and September 2020 in PubMed and EMBASE. The level of evidence A, B, or C and the strength of recommendation (strong or weak) for each recommendation was mapped using the American College of Cardiology/American Heart Association classification system or the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system. The outcome of interest was the proportion of recommendations supported by levels of evidence A, B, and C. Changes in the level of evidence over time were examined. Risk of bias was assessed using Appraisal of Guidelines for Research and Evaluation (AGREE) II. RESULTS: In total, 60 guidelines comprising 2,280 recommendations were reviewed. Level of evidence A supported 16% (363 of 2,280) of total recommendations and 19% (288 of 1,506) of strong recommendations. Level of evidence C supported 51% (1,160 of 2,280) of all recommendations and 50% (756 of 1,506) of strong recommendations. Of all the guidelines, 73% (44 of 60) had a low risk of bias. The proportion of recommendations supported by level of evidence A versus level of evidence C (relative risk ratio, 0.93; 95% CI, 0.18 to 4.74; P = 0.933) or level of evidence B versus level of evidence C (relative risk ratio, 1.63; 95% CI, 0.72 to 3.72; P = 0.243) did not increase in guidelines that were revised. Year of publication was also not associated with increases in the proportion of recommendations supported by level of evidence A (relative risk ratio, 1.07; 95% CI, 0.93 to 1.23; P = 0.340) or level of evidence B (relative risk ratio, 1.05; 95% CI, 0.96 to 1.15; P = 0.283) compared to level of evidence C. CONCLUSIONS: Half of the recommendations in anesthesiology clinical practice guidelines are based on a low level of evidence, and this did not change over time. These findings highlight the need for additional efforts to increase the quality of evidence used to guide decision-making in anesthesiology.

The Adult Cardiac Anesthesiology Section of STS Adult Cardiac Surgery Database: 2020 Update on Quality and Outcomes.

The Society of Cardiovascular Anesthesiologists, in partnership with The Society of Thoracic Surgeons, has developed the Adult Cardiac Anesthesiology Section of the Adult Cardiac Surgery Database. The goal of this landmark collaboration is to advance clinical care, quality, and knowledge, and to demonstrate the value of cardiac anesthesiology in the perioperative care of cardiac surgical patients. Participation in the Adult Cardiac Anesthesiology Section has been optional since its inception in 2014 but has progressively increased. Opportunities for further growth and improvement remain. In this first update report on quality and outcomes of the Adult Cardiac Anesthesiology Section, we present an overview of the clinically significant anesthesia and surgical variables submitted between 2015 and 2018. Our review provides a summary of quality measures and outcomes related to the current practice of cardiothoracic anesthesiology. We also emphasize the potential for addressing high-impact research questions as data accumulate, with the overall goal of elucidating the influence of cardiac anesthesiology contributions to patient outcomes within the framework of the cardiac surgical team.

National Quality Forum Guidelines for Evaluating the Scientific Acceptability of Risk-adjusted Clinical Outcome Measures: A Report From the National Quality Forum Scientific Methods Panel.

: Quality measurement is at the heart of efforts to achieve high-quality surgical and medical care at a lower cost. Without accurate quality measures, it is not possible to appropriately align incentives with quality. The aim of these National Quality Forum (NQF) guidelines is to provide measure developers and other stakeholders with guidance on the standards used by the NQF to evaluate the scientific acceptability of performance measures. Using a methodologically rigorous and transparent process for evaluating health care quality measures is the best insurance that alternative payment plans will truly reward and promote higher quality care. Performance measures need to be credible in order for physicians and hospitals to willingly partner with payers in efforts to improve population outcomes. Our goal in creating this position paper is to promote the transparency of NQF evaluations, improve the quality of performance measurements, and engage surgeons and all other stakeholders to work together to advance the science of performance measurement.

Measuring Childbirth Outcomes Using Administrative and Birth Certificate Data.

BACKGROUND: The number of pregnancy-related deaths and severe maternal complications continues to rise in the United States, and the quality of obstetrical care across U.S. hospitals is uneven. Providing hospitals with performance feedback may help reduce the rates of severe complications in mothers and their newborns. The aim of this study was to develop a risk-adjusted composite measure of severe maternal morbidity and severe newborn morbidity based on administrative and birth certificate data. METHODS: This study was conducted using linked administrative data and birth certificate data from California. Hierarchical logistic regression prediction models for severe maternal morbidity and severe newborn morbidity were developed using 2011 data and validated using 2012 data. The composite metric was calculated using the geometric mean of the risk-standardized rates of severe maternal morbidity and severe newborn morbidity. RESULTS: The study was based on 883,121 obstetric deliveries in 2011 and 2012. The rates of severe maternal morbidity and severe newborn morbidity were 1.53% and 3.67%, respectively. Both the severe maternal morbidity model and the severe newborn models exhibited acceptable levels of discrimination and calibration. Hospital risk-adjusted rates of severe maternal morbidity were poorly correlated with hospital rates of severe newborn morbidity (intraclass correlation coefficient, 0.016). Hospital rankings based on the composite measure exhibited moderate levels of agreement with hospital rankings based either on the maternal measure or the newborn measure (κ statistic 0.49 and 0.60, respectively.) However, 10% of hospitals classified as average using the composite measure had below-average maternal outcomes, and 20% of hospitals classified as average using the composite measure had below-average newborn outcomes. CONCLUSIONS: Maternal and newborn outcomes should be jointly reported because hospital rates of maternal morbidity and newborn morbidity are poorly correlated. This can be done using a childbirth composite measure alongside separate measures of maternal and newborn outcomes.

Impact of the Choice of Risk Model for Identifying Low-risk Patients Using the 2014 American College of Cardiology/American Heart Association Perioperative Guidelines.

WHAT WE ALREADY KNOW ABOUT THIS TOPIC: WHAT THIS ARTICLE TELLS US THAT IS NEW: BACKGROUND:: The 2014 American College of Cardiology Perioperative Guideline recommends risk stratifying patients scheduled to undergo noncardiac surgery using either: (1) the Revised Cardiac Index; (2) the American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator; or (3) the Myocardial Infarction or Cardiac Arrest calculator. The aim of this study is to determine how often these three risk-prediction tools agree on the classification of patients as low risk (less than 1%) of major adverse cardiac event. METHODS: This is a retrospective observational study using a sample of 10,000 patient records. The risk of cardiac complications was calculated for the Revised Cardiac Index and the Myocardial Infarction or Cardiac Arrest models using published coefficients, and for the American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator using the publicly available website. The authors used the intraclass correlation coefficient and kappa analysis to quantify the degree of agreement between these three risk-prediction tools. RESULTS: There is good agreement between the American College of Surgeons National Surgical Quality Improvement Program and Myocardial Infarction or Cardiac Arrest estimates of major adverse cardiac events (intraclass correlation coefficient = 0.68, 95% CI: 0.66 to 0.70), while only poor agreement between (1) American College of Surgeons National Surgical Quality Improvement Program Surgical Risk Calculator and the Revised Cardiac Index (intraclass correlation coefficient = 0.37; 95% CI: 0.34 to 0.40), and (2) Myocardial Infarction or Cardiac Arrest and Revised Cardiac Index (intraclass correlation coefficient = 0.26; 95% CI: 0.23 to 0.30). The three prediction models disagreed 29% of the time on which patients were low risk. CONCLUSIONS: There is wide variability in the predicted risk of cardiac complications using different risk-prediction tools. Including more than one prediction tool in clinical guidelines could lead to differences in decision-making for some patients depending on which risk calculator is used.

Variability in Case Durations for Common Surgical Procedures.

BACKGROUND: Under the Merit-based Incentive Payment System, physician payment will be adjusted using a composite performance score that has 4 components, one of which is resource use. The objective of this exploratory study is to quantify the facility-level variation in surgical case duration for common surgeries to examine the feasibility of using surgical case duration as a performance metric. METHODS: We used data from the National Anesthesia Clinical Outcomes Registry on 404,987 adult patients undergoing one of 6 general surgical or orthopedic procedures: laparoscopic appendectomy, laparoscopic cholecystectomy, laparoscopic cholecystectomy with intraoperative cholangiogram, knee arthroscopy, laminectomy, and total hip replacement. We constructed separate mixed-effects multivariable time-to-event models (survival analysis) for each of the 6 procedures to model surgical case duration. RESULTS: We identified performance outliers, based on surgical case duration, using 2013 data and then quantified the gap between high- and low-performance outliers using 2014 data. After adjusting for patient risk, patients undergoing surgery at high-performance facilities were between 54% and 79% more likely to exit the operating room (OR) per unit time compared to average-performing facilities, depending on the procedure. For example, patients undergoing a laparoscopic appendectomy at high-performance facilities were 68% more likely to exit the OR per unit time (hazard ratio, 1.68; 95% CI, 1.40-2.02; P < .001) compared to average-performing facilities. Patients undergoing a laparoscopic appendectomy at low-performance facilities were 41% less likely to exit the OR per unit time (hazard ratio, 0.59; 95% CI, 0.47-0.74; P < .001) compared to average-performing facilities. The adjusted median surgical case duration for patients undergoing laparoscopic appendectomy was 69 minutes at high-performance centers and 92 minutes at low-performance centers. Similar results were obtained for the other procedures. CONCLUSIONS: There was wide variation in surgery case duration for patients undergoing common general surgical and orthopedic surgeries. This variability in care delivery may represent an important opportunity to promote more efficient use of health care resources.

Impact of Medicare's Nonpayment Program on Hospital-acquired Conditions.

BACKGROUND: Medicare's Nonpayment Program of 2008 (hereafter called Program) withholds hospital reimbursement for costs related to hospital-acquired conditions (HACs). Little is known whether a hospital's Medicare patient load [quantified by the hospital's Medicare utilization ratio (MUR), which is the proportion of inpatient days financed by Medicare] influences its response to the Program. OBJECTIVE: To determine whether the Program was associated with changes in HAC incidence, and whether this association varies across hospitals with differential Medicare patient load. RESEARCH DESIGN: Quasi-experimental study using difference-in-differences estimation. Incidence of HACs before and after Program implementation was compared across hospital MUR quartiles. SUBJECTS: A total of 867,584 elderly Medicare stays for acute myocardial infarction, congestive heart failure, pneumonia, and stroke that were discharged from 159 New York State hospitals from 2005 to 2012. MEASURES: For descriptive analysis, hospital-level mean HAC rates by month, MUR quartile, and Program phase are reported. For multivariate analysis, primary outcome is incidence of the any-or-none indicator for occurrence of at least 1 of 6 HACs. Secondary outcomes are the incidence of each HAC. RESULTS: The Program was associated with decline in incidence of (i) any-or-none indicator among MUR quartile 2 hospitals (conditional odds ratio=0.57; 95% confidence interval, 0.38-0.87), and (ii) catheter-associated urinary tract infections among MUR quartile 3 hospitals (conditional odds ratio=0.30; 95% confidence interval, 0.12-0.75) as compared with MUR quartile 1 hospitals. Significant declines in certain HACs were noted in the stratified analysis. CONCLUSIONS: The Program was associated with decline in incidence of selected HACs, and this decline was variably greater among hospitals with higher MUR.

Impact on hospital ranking of basing readmission measures on a composite endpoint of death or readmission versus readmissions alone.

BACKGROUND: Readmission penalties are central to the Centers for Medicare and Medicaid Services (CMS) efforts to improve patient outcomes and reduce health care spending. However, many clinicians believe that readmission metrics may unfairly penalize low-mortality hospitals because mortality and readmission are competing risks. The objective of this study is to compare hospital ranking based on a composite outcome of death or readmission versus readmission alone. METHODS: We performed a retrospective observational study of 344,565 admissions for acute myocardial infarction (AMI), congestive heart failure (CHF), or pneumoniae (PNEU) using population-based data from the New York State Inpatient Database (NY SID) between 2011 and 2013. Hierarchical logistic regression modeling was used to estimate separate risk-adjustment models for the (1) composite outcome (in-hospital death or readmission within 7-days), and (2) 7-day readmission. Hospital rankings based on the composite measure and the readmission measure were compared using the intraclass correlation coefficient and kappa analysis. RESULTS: Using data from all AMI, CHF, and PNEU admissions, there was substantial agreement between hospital adjusted odds ratio (AOR) based on the composite outcome versus the readmission outcome (intraclass correlation coefficient [ICC] 0.67; 95% CI: 0.56, 0.75). For patients admitted with AMI, there was moderate agreement (ICC 0.53; 95% CI: 0.41, 0.62); for CHF, substantial agreement (ICC 0.72; 95% CI: 0.66, 0.78); and for PNEU, substantial agreement (ICC 0.71; 95% CI: 0.61, 0.78). There was moderate agreement when the composite and readmission metrics were used to classify hospitals as high, average, and low-performance hospitals (κ = 0.54, SE = 0.050). For patients admitted with AMI, there was slight agreement (κ = 0.14, SE = 0.037) between the two metrics. CONCLUSIONS: Hospital performance on readmissions is significantly different from hospital performance on a composite metric based on readmissions and mortality. CMS and policy makers should consider re-assessing the use of readmission metrics for measuring hospital performance.

Impact of Risk Adjustment for Socioeconomic Status on Risk-adjusted Surgical Readmission Rates.

OBJECTIVE: To assess whether differences in readmission rates between safety-net hospitals (SNH) and non-SNHs are due to differences in hospital quality, and to compare the results of hospital profiling with and without SES adjustment. BACKGROUND: In response to concerns that quality measures unfairly penalizes SNH, NQF recently recommended that performance measures adjust for socioeconomic status (SES) when SES is a risk factor for poor patient outcomes. METHODS: Multivariate regression was used to examine the association between SNH status and 30-day readmission after major surgery. The results of hospital profiling with and without SES adjustment were compared using the CMS Hospital Compare and the Hospital Readmissions Reduction Program (HRRP) methodologies. RESULTS: Adjusting for patient risk and SES, patients admitted to SNHs were not more likely to be readmitted compared with patients in in non-SNHs (AOR 1.08; 95% CI:0.95-1.23; P = 0.23). The results of hospital profiling based on Hospital Compare were nearly identical with and without SES adjustment (ICC 0.99, κ 0.96). Using the HRRP threshold approach, 61% of SNHs were assigned to the penalty group versus 50% of non-SNHs. After adjusting for SES, 51% of SNHs were assigned to the penalty group. CONCLUSIONS: Differences in surgery readmissions between SNHs and non-SNHs are due to differences in the patient case mix of low-SES patients, and not due to differences in quality. Adjusting readmission measures for SES leads to changes in hospital ranking using the HRRP threshold approach, but not using the CMS Hospital Compare methodology. CMS should consider either adjusting for the effects of SES when calculating readmission thresholds for HRRP, or replace it with the approach used in Hospital Compare.

Outcomes are Worse in US Patients Undergoing Surgery on Weekends Compared With Weekdays.

BACKGROUND: Increasing surgical access to previously underserved populations in the United States may require a major expansion of the use of operating rooms on weekends to take advantage of unused capacity. Although the so-called weekend effect for surgery has been described in other countries, it is unknown whether US patients undergoing moderate-to-high risk surgery on weekends are more likely to experience worse outcomes than patients undergoing surgery on weekdays. OBJECTIVE: The aim of this study was to determine whether patients undergoing surgery on weekends are more likely to die or experience a major complication compared with patients undergoing surgery on a weekday. RESEARCH DESIGN: Using all-payer data, we conducted a retrospective cohort study of 305,853 patients undergoing isolated coronary artery bypass graft surgery, colorectal surgery, open repair of abdominal aortic aneurysm, endovascular repair of abdominal aortic aneurysm, and lower extremity revascularization. We compared in-hospital mortality and major complications for weekday versus weekend surgery using multivariable logistic regression analysis. RESULTS: After controlling for patient risk and surgery type, weekend elective surgery [adjusted odds ratio (AOR)=3.18; 95% confidence interval (CI), 2.26-4.49; P<0.001] and weekend urgent surgery (AOR=2.11; 95% CI, 1.68-2.66; P<0.001) were associated with a higher risk of death compared with weekday surgery. Weekend elective (AOR=1.58; 95% CI, 1.29-1.93; P<0.001) and weekend urgent surgery (AOR=1.61; 95% CI, 1.42-1.82; P<0.001) were also associated with a higher risk of major complications compared with weekday surgery. CONCLUSIONS: Patients undergoing nonemergent major cardiac and noncardiac surgery on the weekends have a clinically significantly increased risk of death and major complications compared with patients undergoing surgery on weekdays. These findings should prompt decision makers to seek to better understand factors, such physician and nurse staffing, which may contribute to the weekend effect.

Quality of Quality Measurement: Impact of Risk Adjustment, Hospital Volume, and Hospital Performance.

BACKGROUND: The validity of basing healthcare reimbursement policy on pay-for-performance is grounded in the accuracy of performance measurement. METHODS: Monte Carlo simulation was used to examine the accuracy of performance profiling as a function of statistical methodology, case volume, and the extent to which hospital or physician performance deviates from the average. RESULTS: There is extensive variation in the true-positive rate and false discovery rate as a function of model specification, hospital quality, and hospital case volume. Hierarchical and nonhierarchical modeling are both highly accurate at very high case volumes for very low-quality hospitals. At equivalent case volumes and hospital effect sizes, the true-positive rate is higher for nonhierarchical modeling than for hierarchical modeling, but the false discovery rate is generally much lower for hierarchical modeling than for nonhierarchical modeling. At low hospital case volumes (200) that are typical for many procedures, and for hospitals with twice the rate of death or major complications for patients undergoing isolated coronary artery bypass graft surgery at the average hospital, hierarchical modeling missed 90.6% of low-quality hospitals, whereas nonhierarchical modeling missed 65.3%. However, at low case volumes, 38.9% of hospitals classified as low-quality outliers using nonhierarchical modeling were actually average quality, compared to 5.3% using hierarchical modeling. CONCLUSIONS: Nonhierarchical modeling frequently misclassified average-quality hospitals as low quality. Hierarchical modeling commonly misclassified low-quality hospitals as average. Assuming that the consequences of misclassifying an average-quality hospital as low quality outweigh the consequences of misclassifying a low-quality hospital as average, hierarchical modeling may be the better choice for quality measurement.